Metraipolis › archive › dc_16dee3f4be51

Segment What It Can Draw: Querying a Text-to-Image Generator with Its Own Images for Skin Lesion Segmentation — arXiv:2611.07061

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.07061
License (as archived)
CC0-1.0
Archived
2026-11-06T14:42:08.948Z
SHA-256 of the archived text
c1b14858fd120a6b5468d94dc792f6a402001d58b2efd4bc6270504524028ed1
Stable URL of this page
https://metraipolis.com/documento/dc_16dee3f4be51

How to cite

Segment What It Can Draw: Querying a Text-to-Image Generator with Its Own Images for Skin Lesion Segmentation — arXiv:2611.07061 — arXiv (en), https://arxiv.org/abs/2611.07061 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_16dee3f4be51 (Metraipolis archive dc_16dee3f4be51, sha256 c1b14858fd120a6b5468d94dc792f6a402001d58b2efd4bc6270504524028ed1)
BibTeX
@misc{dc16dee3f4be51,
  title = {Segment What It Can Draw: Querying a Text-to-Image Generator with Its Own Images for Skin Lesion Segmentation — arXiv:2611.07061},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.07061},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_16dee3f4be51; sha256 c1b14858fd120a6b5468d94dc792f6a402001d58b2efd4bc6270504524028ed1},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.07061. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Skin lesion segmentation without annotated masks requires the target to be specified through other cues. Existing approaches rely on dermoscopy-specific rules, annotation-derived model priors, or user-selected regions. A general text-to-image diffusion transformer offers a text interface to the target, but its concept-word attention is unreliable on dermoscopy images: in the model's attention-output space, the word-derived skin-to-lesion direction is poorly aligned with the direction separating real skin from real lesions. We show that the direction derived from generated healthy-skin and lesion images is substantially better aligned. Based on this finding, we propose GENIQ, a training-free method that constructs initial skin and lesion prototypes from generated images and refines them using the most skin-like and lesion-like patches from unlabelled dermoscopy images. It scores test patches along the refined direction and applies a gated threshold to suppress contextual spill-over around small lesions. Without inference-time annotations or additional models, GENIQ achieves Dice scores of 0.85-0.88 on three dermoscopy test sets. It outperforms zero-shot and word-based attention methods on all three sets, and a hand-designed dermoscopy pipeline in Dice and IoU on two.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2611.07061 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-06T14:18:42.665Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).